btw the same happens when I try this on localhost
# ClearML SDK configuration file api { # Notice: 'host' is the api server (default port 8008), not the web server. api_server:
web_server:
files_server:
# Credentials are generated using the webapp,
# Override with os environment: CLEARML_API_ACCESS_KEY / CLEARML_API_SECRET_KEY
Yes other then the link it generates it works fine
And all the others works as expected?
GreasyPenguin14 I was able to reproduce it, will keep you update when a fix is ready
I enqueue to service to the services queue, not done anything myself with agents
Are you running the services with an agent?
Which version do you use?
Hi GreasyPenguin14 ,
The reason for this is the Services Agent is not aware of the WebApp address, and tries to guess by using the ApiServer address (which internally, in docker-compose, is http://apiserver:8008 ) by replacing the port to 8080
.
To fix it, you simply need to specify the correct address by which the WebApp can be accessed when starting the server using docker-compose
.
The easiest way to do that is to set the environment variable CLEARML_WEB_HOST
before starting docker-compose, in your case:> CLEARML_WEB_HOST=
docker-compose -f /opt/clearml/docker-compose.yml up -d
I created it with clearml-init, nothing special. It looks like
# ClearML SDK configuration file api { # Notice: 'host' is the api server (default port 8008), not the web server. api_server:
web_server:
files_server:
# Credentials are generated using the webapp,
# Override with os environment: CLEARML_API_ACCESS_KEY / CLEARML_API_SECRET_KEY credentials {"access_key": "NMWOE5C3RGTX473M9D3M", "secret_key": "L@G50jO+TJ23#8Eerp1E$4y=elUt11P!BL2Ouas=1$GY-GxY%^"} } sdk { # ClearML - default SDK configuration
I tried it but it doesnot work. I checked the clearml-agent-services container and it does have the right CLEARML_WEB_HOST
echo $CLEARML_WEB_HOST
http://192.168.4.1:8080
however, the messages on slack still point to apiserver:8080
Hi GreasyPenguin14 ,
What is the web server configure in your ~/clearml.conf
file ( api.web_server
)?
It seems there might be a bug there -we'll release an RC soon to fix it